Search

SPIDER is the prototype ion source for ITER Heating Neutral Beam. To investigate the phase-space distribution of its beamlets, an Allison type emittance scanner was developed. A slit in front of the device selects a part of the beam. A uniform electric field perpendicular to the beam, generated polarizing two parallel metallic plates, then deflects the particles. Only those with a well-defined entrance angle pass through another slit, aligned with the first one. The emittance profile of the beam is obtained by moving the device perpendicularly to the beam during the pulses, while sweeping the electric field between the plates and collecting the current reaching the Faraday cup placed after the second slit. This kind of diagnostic is widely used in ion sources for high-energy accelerators, but has not yet been extensively employed in fusion-relevant ion sources. We present in this paper the experimental setup of the diagnostic, together with the proposed operating procedure and the expected operating space. Tests carried on to assess the performance of the device, both with the Faraday cup and with the entire detector are described. The emittance measurements obtained in a dedicated test facility are also reported.

Magnetic fields in negative ion driven Neutral Beam Injectors (NBIs) are essential to reduce the electron temperature and density in the extraction region in order to consequently reduce the destruction rate of H- ions, along with filtering out the otherwise unavoidably co-extracted electrons from the plasma source and beam accelerator. In SPIDER, the full scale ion source and extractor of the ITER NBI, a horizontal magnetic field of some mT inside the plasma source, transverse with respect to the beam direction, is produced by a current of some kA flowing through the plasma-facing electrode and additional bus-bars. After an initial characterization phase, the first experimental campaign with relevant plasma generation and particle extraction showed that when the magnetic field is increased beyond a certain threshold, an influence on the plasma parameters is observed. This effect has been ascribed to the particular topology of the magnetic field in the plasma volume inside the 8 inductively-coupled radio-frequency drivers that generate the plasma. Therefore, actions were immediately taken to modify the magnetic field configuration in order to improve the plasma diffusion out of the drivers. The paper describes the experiments aimed at identifying the problem and the design and optimization process for a new layout of the bus-bars creating the magnetic filter field, in order to obtain a different topology within the region of the RF driver, while keeping it almost unchanged in the other regions where the required performances have already been achieved. Another aim of the work concerns the possibility of easily modifying the new topology, so as to significantly expand the operating margin of the experimental activity. After introducing the reasons of the work, the numerical analysis and experimental verification activities necessary for the development and implementation of the new system are presented.

The SPIDER ion source has to generate a current of H- ions up to 60 A and accelerate them up to 100 keV. In addition, the ratio between the co-extracted electrons and negative ions has to be below 1, while a beam inhomogeneity within 10% has to be attained. The source plasma is generated inside 8 inductively coupled radio-frequency drivers; it then diffuses through a so-called expansion region and finally negative ions are extracted and accelerated by means of a three-grid system comprising a plasma grid (facing the source plasma), an extraction grid and a grounded grid. A magnetic filter field is produced by flowing a current through the plasma grid in the vertical direction with the aim of reducing electron temperature and density in the extraction region so as to enhance the survival probability of H- ions while reducing the co-extracted electrons. The main drawback is that the coupling of the magnetic filter field with the electric field and the pressure gradients within the ion source induces a vertical plasma drift which possibly turns into a vertical beam inhomogeneity. Another strategy to reduce the co-extracted electrons is to bias the plasma grid with respect to the ion source body and to install a so-called bias plate in the vicinity of the plasma grid, which can also be biased. Also, such biases affect the beam inhomogeneity by modifying the ExB drift. After the first beam campaign, the filter field configuration of SPIDER was modified so as to improve the plasma diffusion outside of the drivers. In addition, the voltage ratings of the bias power supplies were increased. Both these modifications permit to lower the electron co-extracted current while increasing the negative ion current. The present contribution describes such improvements while focusing on the effects they have on the beam inhomogeneity.

SPIDER, the full-size prototype of the negative ion source for future ITER Heating and Current Drive, is equipped with a tomographic diagnostic to characterize the negative hydrogen beam. The main goal of the system is the reconstruction of the 2-dimensional beam emission profile, in the plane x-y perpendicular to the beam propagation direction (z), which is proportional to the beam density by assuming a uniform background gas. The light emitted is mainly produced by the deexcitation of fast neutrals produced by the interaction of the beam particles with the background gas, with the emission of a photons with ?=656.28 nm (the Balmer line H?). Eight visible cameras observe the beam at a distance of 50 cm from the last grid, all around the beam: 1 linear camera and 7 cameras equipped with 2D sensors of different sizes and with different lenses, to have a spatial resolution in the order of the single beamlet dimension (centimetre), despite the limited number of point of views. To reconstruct the profile of the extracted beam, the cameras are calibrated in intensity, aligned and focused on the plane perpendicular to the beam propagation direction. Their fans of lines of sight are reconstructed and the experimental data are compared with the simulated ones. This analysis allows to characterize the beam homogeneity, deflection and divergence, which is estimated by assuming Gaussian profiles for the beamlets at different positions along the beam propagation direction, in different experimental conditions. The evolution of these quantities as a function of the main source parameters is presented.

Giant negative-ion sources for fusion plasma heating are optimized to provide, by a peculiar magnetic topology in the expansion region, a separation between a relatively high electron-temperature region for plasma formation and hydrogen dissociation, and a beam extraction region in which low electron-temperature is required to avoid negative-ion destruction. The full-scale ITER neutral beam ion-source prototype SPIDER is the largest RF-driven negative ion source in operation, and it exhibits non-uniformities and drifts in the large expansion region from the eight RF drivers. A direct measurement of the plasma density and electron temperature in the expansion and extraction region as a function of the operational parameters is therefore required to correlate negative-ion beam uniformity to the operational parameters. The main operational parameters affecting plasma expansion and uniformity are the gas pressure, the perpendicular magnetic field strength, and the electric bias of ion source surfaces. This paper describes the development of a set of movable electrostatic probes that are used to measure axial profiles of plasma parameters, entering the ion source from the electrode apertures of the ion accelerator towards the RF drivers. The set of electrostatic probes encompasses eight RF-compensated Langmuir probes, one double probe, one triple probe for relatively high-frequency measurements, one Mach probe for the assessment of plasma drift velocities, and two gridded retarding field energy analyzers to measure the positive-ion energy distribution function. The mechanical constraints given by the experimental setup, based on a large in-vacuum movable structure with multiple probes entering from small apertures and required to scan about 400 mm, the thermal requirements of these relatively heat-flux components, and the electrical aspects, made the design challenging for the limited allowable dimensions. Prototyping and commissioning of the measurement system is also discussed. Examples of measured axial profiles with the various probes used in SPIDER will be presented.

SPIDER (Source for Production of Ions of Deuterium Extracted from Radio frequency) experiment is operating at PRIMA (Padova Research on ITER Megavolt Accelerator) site since June 2018. It is a test bed to study the operation of an ITER-relevant negative ion source. Caesium (Cs) adsorption on the extraction system, in particular on the plasma grid, lowers the surface work function. This increases the probability of negative ions production and the reduction of coextracted electrons, as experimentally observed. The Cs layer in the SPIDER beam source shall be built up in a controlled and repeatable way by injecting Cs vapors by means of three Cs ovens. A SPIDER Cs oven prototype was procured and tested in the CAesium Test Stand (CATS) at PRIMA site validating the design. Following the experimental tests, few improvements were introduced in the final design of the three Cs ovens for the SPIDER source. Then, the ovens were procured and commissioned before the installation in SPIDER. This paper reports the functional tests of the SPIDER Cs ovens and of their embedded diagnostics. These tests were carried out in CATS, with the aim of characterizing the three ovens and of assessing their reproducibility, in order to validate them for use in SPIDER in the planned experimental campaigns with caesium evaporated in the source.

SPIDER is the full-size prototype of ITER Neutral Beam Injector ion source, in operation since June 2018 as a project of the ITER Neutral Beam Test Facility located at Padova, Italy. It consists of an inductively coupled plasma source and an extraction system to accelerate the produced negative ions up to 100 keV energy. The plasma source is composed of 8 drivers, operated in low hydrogen gas pressure (~0.3 Pa) and maximum radio frequency (RF) power of 100 kW per driver at 1 MHz frequency.A key parameter to qualify performances of the driver is the power transfer efficiency(PTE) which is defined as the ratio between the power absorbed by the plasma andthe total RF input power. Experimentally it is not possible to measure the power absorbed by the plasma, which is found to depend on various parameters coupled together. Therefore a model is needed which can account for all the mechanisms toprovide an estimation of this PTE. A methodology was developed in this regard based on the integration of two main models: analytical, to account for different mechanisms of plasma particle dynamic;and electrical, accounting for coil, plasma and passive metallic structures present within the driver region. The methodology was then applied to two different ion sources.This work highlights a further development of this methodology. Two main additions in analytical model are discussed: the estimation of average electron density as a function of RF power and gas pressure; and the influence of the magnetic field on the input parameters for the electrical model (for instance, plasma conductivity). The methodology is then applied to SPIDER drivers and the results in terms of equivalent electrical parameters and PTE are presented.

The Experimental Fusion Reactor ITER will use fast Neutral Beams (NBs) injectors to increase the plasma temperature in order to drive the plasma current necessary for stability. The NBs are produced by D- beams, which are generated by negative ion sources and neutralized by gas cells with an efficiency of 60%. In order to improve the electrical efficiency of the NB production, a beam recovery system of the residual charged particles could be foreseen. In particular, we have already proposed an energy recovery device with an axisymmetric cylindrical ion collector. A scaled ITER type negative ion source, NIO1 (Negative Ion Optimization 1), has been developed at RFX (Padua) for studying and improving the ion extraction efficiency. An experimental test on beam energy recovery by using the NIO1 ion beam has been recently proposed. However, one important issue related to the beam energy recovery are the secondary electrons generated by the interaction of the ions with the collector walls which may escape from the device, reducing energy recovery efficiency and cause damages. In this contribution, the problem of minimizing the secondary electrons production in the energy recovery device has been studied and simulated. Although a fraction of the secondary electrons escapes from the ion collector, they are greatly stopped in the axisymmetric geometry cup. A simple test bench for secondary electron emission to verify the simulation results is in preparation at INFN-BA laboratory.

The instrumented calorimeter STRIKE (Short-Time Retractable Instrumented Kalorimeter Experiment) has been designed with the main purpose of characterizing the SPIDER negative ion beam in terms of beam uniformity and divergence during short pulse operations. STRIKE is made of 16 1D Carbon Fibre Composite (CFC) tiles, intercepting the whole beam and observed on the rear side by infrared (IR) cameras. The front observation presents some drawbacks due to optically emitting layer caused by the excited gas between the beam source and the calorimeter, and the material sublimated from the calorimeter surfaces due to the heating itself. It is then necessary to solve an inverse non-linear problem to determine the energy flux profile impinging on the calorimeter, from the 2D temperature pattern measured on the rear side of the tiles. Most of the conventional methods used to solve this inverse problem are unbearably time consuming, so a ready-to-go instrument to determine the beam condition, while operating STRIKE, is mandatory. In this work, the inverse problem, both in stationary and non-stationary conditions, is faced by using a Neural Network (NN) model, pursuing two different approaches. In the first one, the NN is trained to directly solve the inverse problem, by associating the radiation profile (target) to the measured temperature profile (input). In the second approach, a NN is trained to solve the direct problem, where the input is the radiation profile and the target is the temperature profile. Then, the NN is inverted by determining the input corresponding to a fixed target. Preliminary results show the reliability of the proposed method for STRIKE real time operation.